Crucial gaps in knowledge prevent the providing of ideal individualized care to women with endometrial cancer. The extent of lymphadenectomy and the identification of which patients would benefit the most from complete surgical staging are unresolved clinical issues. A second important gap is the ability to predict which women with low stage endometrioid-type endometrial cancer will suffer recurrence. We have used genomic approaches to identify tissue biomarkers of endometrial cancer. These biomarkers have been validated in a large set of endometrial cancers at MDACC, and we have documented their close association with stage and recurrence. Next, we will validate these findings in independent data sets. We hypothesize that quantifying genes associated with EMT and estrogen's growth regulatory actions in the endometrium as a "biomarker score" will provide a clinically useful tool to assist in the decision to perform complete surgical staging on women diagnosed with endometrial cancer and will predict which women with stage I and stage II endometrioid carcinomas will recur. In Aim 1, we will validate the association of our biomarkers with endometrial cancer stage in an independent data set obtained from a series of patients who unden/vent comprehensive surgical staging at the Mayo clinic. We will also establish that the biomarker score computed from formalin-fixed, paraffin-embedded endometrial biopsies is a good approximation of the same scores based on tissue from the final hysterectomy. We will determine the association of the biomarker panel with endometrioid carcinoma recurrence in independent data sets obtained from Mayo Clinic and Washington University. In Aim 2, reverse phase protein lysate array (RPPA) will be used to identify proteins and phospho-proteins that are associated with endometrioid carcinoma stage. These newly identified biomarkers will be used to augment our currently existing panel. We will also develop methodology to perform RPPA using formalin-fixed tissues so that this technology becomes more relevant to clinical samples. Finally RPPA will be used to help identify proteins that interact with EIG121, one of the more exciting but least understood biomarkers in our existing panel.